The problem of analog circuit design is a difficult problem thatis generally viewed as requiring human intelligence to solve.Considerable progress has been made in automating the design of certaincategories of purely digital circuits; however, the design of analogelectrical circuits and mixed analog-digital circuits has not proved tobe as amenable to automation. When critical analog circuits are requiredfor a project, skilled and highly trained experts are necessary.Previous work on applying genetic programming to the design of analogcircuits has proved to be successful at evolving a wide variety ofcircuits, including filters, amplifiers and computational circuits;however, these previous approaches have required the specification of anappropriate embryonic circuit. This paper explores a method to eliminateeven this small amount of problem-specific knowledge, and, in addition,proves that the representation used is capable of producing all circuits
展开▼